Patterns of Argumentation in Gorgias‘ΕΦΦ263/01 Patterns of... · of argumentation is not a...
Transcript of Patterns of Argumentation in Gorgias‘ΕΦΦ263/01 Patterns of... · of argumentation is not a...
Patterns of Argumentation in GorgiasAuthor(s): D. G. SpatharasReviewed work(s):Source: Mnemosyne, Fourth Series, Vol. 54, Fasc. 4 (Aug., 2001), pp. 393-408Published by: BRILLStable URL: http://www.jstor.org/stable/4433237 .Accessed: 02/04/2012 04:55
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
BRILL is collaborating with JSTOR to digitize, preserve and extend access to Mnemosyne.
http://www.jstor.org
PATTERNS OF ARGUMENTATION IN GORGIAS*
BY
D. G. SPATHARAS
Gorgias, according to Plato's Phaedros (267a), valued probabilities
(e???ta) ahead of truth; in other words, Gorgias preferred proba- bilities to actual truth or?in a morally marked modification of this
disclaimer?truth was practically neglected by him. To put it in
Dodds' words he was the producer of "dazzling insincerities". An
interesting version of this standard idea about Gorgias' argumenta- tion is the one offered by Kerferd; commenting on The Encomium
of Helen he says that Gorgias' "emphasis on truth" is "emphatic", because "deceit is only possible in relation to that which is actually
true", which entails that truth in Gorgias' hands is nothing but a
presupposition of effective deceit.1) This view has been challenged
only very recently by Gagarin, who, instead of reading Plato's eval-
uation of early rhetoric, decided to read Gorgias' himself (and other
early masters of rhetoric) and he concluded that what we have from
Plato and some modern scholars is a distorted image of the so-
called early rhetoric.2) In this paper, following Gagarin's paradigm, I wish to shed some
more light on Gorgias' argumentation; by attempting a classification
of argumentative patterns employed by this Sophist, it will become
more clear that arguments from probability is only one among sev-
eral other patterns of reasoning used by him, and consequently that
Plato singles out this argumentative pattern, because probabilities
* I am grateful to Prof. D. M. MacDowell who read an earlier draft of this paper, and to Thomas McGrory who improved my English significantly.
1) E. R. Dodds, Plato, Gorgias (Oxford 1959), 8; G. B. Kerferd, The Sophistic Movement (Cambridge 1981), 79. On modern scholars and argument from proba- bilities, see M. Gagarin, Probability and Persuasion, in: I. Worthington (ed.), Persuasion: Greek Rhetoric in Action (London 1994), 49-50.
2) In his own words, "Plato's criticisms on this point reflect his own concern with the overriding primacy of an absolute standard of truth, which is tied to and validated by his Forms; for him anything less than absolute truth was no truth at all", Gagarin (n. 1), 56-57.
? Koninklijke Brill NV, Leiden, 2001 Mnemosyne, Vol. LIV, Fase. 4
394 D. G. SPATHARAS
can easily be regarded as opposed to factual truth. Moreover, I pro-
pose to show that Gorgias' process of argumentation is more com-
plicated than it is usually taken to be. The patterns I wish to examine
seem to be relatively recurrent in Gorgias' preserved texts (that is
The Encomium of Helen, The Defence of Palamedes, and On not Being;3) I
exclude the Epitaphios which should be considered as a longer frag-
ment). They are the following: 1) arguments from probabilities, 2)
arguments from antinomy, 3) 'theorisation' through examples, and
finally 4) apagogie and 'Russian doll' argumentation.
1. Arguments from probabilities
The term argument from probabilities (e????) means an argument which is not based on definitive factual reality; it is an argument the acceptability of which depends on its potential to reproduce facts on the grounds of common experience shared by humans. It
is not necessary to go into details concerning the history of the
probability arguments in Greek literature.4) It will suffice to men-
tion the interesting case of the "reverse probability" argument (to use Gagarin's words), attributed by Aristotle (Rhet. 1402a 17-28) to
Corax; in a conflict between a weak and a strong man the weak
man can simply argue that he did not attack the strong man because
he was not able to do so; this argument cannot be employed by the strong man who can instead use the reverse probability: if I
had attacked the weak man the obvious suspect would have been
myself, so I never attacked him.
As far as I know, the reverse probability is never used by Gorgias,
although his Palamedes makes an ample use of e???ta;5) according
3) Unfortunately Gorgias' original text has not been preserved; all the infor- mation we possess on its content is due to the accounts of the Anonymus author of the treatise De Melisso Xenophane Gorgia (MXG) 5-6, 979al2-980b21, and that of Sextus Empiricus' Adver sus mathematicos 7,65-87.
4) Its origins are traditionally placed with Corax and Teisias, but it can be traced in early literature (see Gagarin (n. 1), 51, who maintains that "they did, however, develop a new form of this argument?.. . the 'reverse probability' argu- ment"). On Corax and Teisias see S. Wilcox, Corax and the Prolegomena, AJP 64
(1943), 1-23; D. A. G. Hinks, Teisias and Corax and the Invention of Rhetoric, CQ, 34
(1940), 61-69; and T. Cole, Who Was Corax?, ICS 16 (1991), 65-84.
5) The argument from probabilities in Gorgias has been discussed by I. Anastassiou, 7/ ???a??????s? ?a? ????d?? ?p?de???? st? G????a, in ?. ?. Voudouris (ed.), ? '???a?a S?f?st??? (Athens 1982), 242-250.
PATTERNS OF ARGUMENTATION IN GORGIAS 395
to the version of the myth followed by Gorgias,6) Palamedes was
falsely accused by Odysseus that he betrayed Greece to the Trojans. In order to support this accusation, he forged a letter ostensibly sent by Priam to Palamedes and placed it in the tents of the lat-
ter along with a quantity of gold. When Odysseus 'found' the gold and the letter, he accused Palamedes of treason, and after a trial
heard by the Greek leaders the wise hero was condemned to death.
Gorgias divides his defence into two major parts: a) if I had wanted
to betray the Greeks to the barbarians I would not have been able
to do so (6-12), and b) if I had been able to do it I would not
have wanted to do it (13-21).7) The development of the argumen- tation in both parts makes abundant use of probabilities: in the first
part, Palamedes proves that all the necessary stages for the prepa- ration of the betrayal (communication with Priam, a pledge, the
transfer of money etc.) were all impossible. His argumentation pro- ceeds by conceding each step: in order to betray Greece I had to
do A which was impossible; but even if A had been possible ?
would have been necessary; but ? was impossible; even if ? had
been possible, and so on.8) This first part is, basically, a presenta- tion of probabilities which are proved to be invalid, due to practi- cal reasons. In the second part, probabilities are employed in relation
to motives. It is an exhaustive examination of possible reasons for
which one might have been tempted to commit the crime of trea-
son (money, power, helping friends etc.); each one of these motives
is dealt with separately and much of the persuasiveness lies both in
the conformity of the hero with generally accepted moral standards
and in the detailed discussion of practically every possible motive.
The Encomium of Helen is not marked by the use of probabili- ties; the mythological version adopted here is the Homeric one,
according to which Helen did travel to Troy with Paris (f. 5 ep?a?e?
6) For Palamedes in Greek Literature see the most detailed, but less quoted G. Z. Lyra, ? ????? t?? ?a?a??d? st?? '???a?a ???????? G?a??ate?a (Yiannina 1987); see also R. Scodel, The Trojan Tribgy of Euripides (Hypomnemata 60; G?ttingen 1980), 43-63.
7) Notice that when probability arguments are employed, hypothetical clauses are very likely to appear; an e???? must have as its starting point a hypothesis, which is either confirmed or rejected; in Palamedes the two hypotheses are rejected, whereas in Helen the validity of the four hypothetical reasons for which Helen deserted her home is confirmed (see also Anastassiou (n. 5), 244).
8) See below, 'the Russian doll' argumentation.
396 D. G. SPATHARAS
a ep?a?e?); what is questioned in this speech is the responsibility of
Helen.9) The invention of the reasons which made it possible for
Helen to travel to Troy is the only trace of probabilities in this
speech (cp. 6 ta? a?t?a?, d?' a? e???? rjv ?e??s?a? t?? t?? ?????? e??
t?? ????a? st????). These reasons are the following: a) the wish of
the gods, b) natural violence, c) speech-persuasion, and d) love. Each
reason is dealt with separately (as are the motives in Palamedes)?) and they are not mutually exclusive. At any rate, Helen is inno-
cent; but her innocence, in this case, does not mean that she did
not desert her husband, or that she did not travel to Troy with
another man: it means lack of personal responsibility. The absence of probability-arguments in Helen, and the ample
use of them in Palamedes is a first indication that Gorgias did not
actually prefer probabilities to truth, and that the use of this type of argumentation is not a matter of choice, but a matter of neces-
sity.11) In the case of Palamedes, Gorgias undertakes the defence of
a man accused falsely. At the very beginning of the speech (4), he
has Palamedes claim that he is in a situation of ap???a (although he is p?????? par excellence 25, and indeed the hero who made human
life p?????? ?? ap???? 30) due to e?p????? created by groundless accusation. The only thing Palamedes can do is trust 'truth' (a???e?a),
9) T. Cole, The Origins of Greek Rhetoric (London 1991), 76 claims that Helen "is an illustration of what later rhetoricians (for example, Quint. 7.41) would call the status qualitativus"', whereas "Palamedes' defense ... provides a model for the status coniecturalis?the type of argument concerned with determining what actually occurred (cf. Cicero, De inv. 1.8.10)". M. Gagarin, Antiphon, The Speeches (Cambridge 1997), 122 saw both in Gorgias' Helen and Palamedes and in Antiphon's Tetralogies "a fore- shadowing of stasis-theory in Aristotle {Rhet. 1.13.9-10, 1373b38-74al7), who dis-
tinguishes cases where the facts are in dispute from those where the facts are admitted but the defendant denies there is a crime". At any rate, Gorgias is aware of the fact that the two cases differ in this respect, and this awareness is depicted in both texts: Helen ??? d? ??????a?... ??a??a? 2, p????s??a? ta? a?t?a?, d?' a? e???? ?? ?e??s?a? t?? t?? ?????? e?? ????a? st???? 5, and Palamedes ??d? ??d' dp?? a? e?de?? t?? d? t? ?? ?e???e??? 5.
10) See below, 'apagogie' argumentation. 11) Gagarin (n. 1), 54 rightly observes that Helen's "case seems well suited for
giving probability a higher value than truth, since many different versions of Helen's actions existed ... in which she did not go to Troy", and concludes "Gorgias has no reason to resort to probability arguments, since the basic facts are known and
accepted", though it is not certain if Euripides' Helen (included by Gagarin in these versions) existed before Gorgias' Helen. The dates of both Palamedes and Helen are uncertain; for some conjectures see M. Orsini, La cronologia dell' "Encomio di Elena" di Gorgia e le "Troiane" di Euripide, Dioniso 19 (1956), 82-88 with summary.
PATTERNS OF ARGUMENTATION IN GORGIAS 397
and 'compulsion' (a?????), which (or who) are dangerous rather
than resourceful teachers (d?das????? ep????d???te??? ? p?????t????).
Why is truth a dangerous teacher? Because in his case, the truth
is that he has not committed the crime that he is accused of?a
disclaimer which admittedly is not of much convincing value?espe-
cially in the absence of witnesses. Gorgias certainly knows that judges are not persuaded by mere statements of innocence, which entails
that an approximation of truth through arguments should be employed instead. This is exacdy what probabilities serve for. In the case of
Helen, the defence does not refute the facts; Gorgias, neglecting the
morale of Stesichorus' suffering, seems confident enough to adopt the Homeric version of the myth: Helen did go to Troy. By doing
so, he does not need probabilities. What is at stake in Helen is the
removal of her infamy on the grounds of reasonable excuses; the
notorious trouble-maker has been the victim of uncontrolled powers. It seems thus, that probabilities appear when facts are disputed
and that they are more of a necessity than mere choice; but even
in the discussion of factual reality, Gorgias does not always prefer
probabilities. He must have realized that other means of persuasion are sometimes equally effective and convenient. In Helen 13, he
claims that a speech written with skill (t????) persuades an audi-
ence without necessarily telling the truth. The form of a speech is
there considered as a determinant factor of persuasiveness; if the
admirers of truth may now feel ready to argue that this is a fur-
ther proof of the (conscious) sacrifice of real facts for the sake of
persuasion, there is the counter-argument that in cases like the one
of Palamedes, the truth is that he is an innocent person, who has
the serious task to defend his own honour and life. Who, in other
words, would today accuse an advocate defending a victim of con-
spiracy on the grounds of his/her eloquence, especially in a case
where the evidence is lacking? Another vehicle of persuasion is, of course, direct evidence given
by witnesses. In Palamedes (22) the defendant holds that an accusa-
tion is stronger, when it is accompanied by witnesses (p?st?te???
?a? ??t?? t? ?at??????a ?a?t??????).12) In the following paragraph
(23) the hero addresses Odysseus with the following words:
12) The false accusation of Odysseus is implied in Palamedes 7, where the defen- dant, for the argument's sake, concedes the possibility of communication with
398 D. G. SPATHARAS
???a s?? ??? ??? ?? ???? (te) ????? ???t??a? ???a ?a? ?e?d????t??a? e??e??, ???? d? ??d?te??? e??e?? t??t?? d??at??.
The point is that Palamedes, being innocent, could not find any witnesses of a crime which has never been committed; but Odysseus, on the contrary, was able to present both witnesses?in case Palamedes
has committed the crime?and false-witnesses13)?as, of course, he has
not committed the crime. The reference to witnesses, that is to say
people with personal knowledge, shows that Gorgias was aware of
and able to use means which normally furnish direct evidence. This
point becomes more interesting, when Gorgias attempts to trans-
form Odysseus from an accuser into a witness or even an accom-
plice (22):
e? ??? ?a? e?d??, ??s?a ?d?? ? ?et???? ? t?? (?et????t??) p????e???. e? ??? ??? ?d??, f??s?? t??t??? (t?? t??p??), t?? t?p??, t?? ??????, p?te, p??, p?? e?de?* e? d? ?et????, ?????? e?? ta?? a?ta?? a?t?a?? . . .
To sum up, although probabilities are an important vehicle of argu-
mentation, Gorgias does not value them ahead of truth; e???ta are
necessary in cases where real facts are disputed and where evidence
is lacking. Pleasure invoked in the audience by a skilled speech is
also regarded by Gorgias as a factor which plays an important role
in its persuasiveness, so that a good rhetor should take this para- meter into account. Last?y, it is clear that direct evidence presented
by witnesses is known to Gorgias, and, what is more, he seems to
be aware of the function of this type of evidence, so that he can
argue by using it.
2. Argument from antinomy
This pattern of argumentation (Aristode describes it as t?p?? ?? t??
e?a?t???) is founded on the location of antinomic or contradictory
properties attributed to one and the same entity.14) It occurs both
Priam, which entails that an interpreter should have been used. If this had occurred, the latter would have been a witness of the transaction. The failure of the oppo- nent to provide witnesses or to accept evidence by means of ??sa??? (cp. Palamedes
11) is a topos*, see Antiphon 1.6.13, 23 and 29-30. Most recently I. M. Plant, The
Influence of Forensic Oratory on Thucydides' Principles of Method, CQ49 (1999), 66-67, 71.
13) For ?e?d??a?t???a see for example Antiphon 2.4.7, Andokides 1.7, Lysias 19.4.
14) See G. E. R. Lloyd, Polarity and Analogy (Cambridge 1971), 121 and J. Mansfeld, Historical and Philosophical Aspects of Gorgias' "On What Is Not", in: L. Montoneri -
PATTERNS OF ARGUMENTATION IN GORGIAS 399
in On not Bang (OMB), where it is used for the refutation of philo-
sophical arguments and in Palamedes, where it is integrated in the
characterization of the opponent. In OMB Gorgias puts forward three major theses: a) nothing is,
b) if it is it is unknowable, and c) if it is and it is knowable it can-
not be communicated to others. A problem which has tantalized
scholars is what exacdy this 'it5 refers to. Some scholars have said
that it is the phenomenal world in general and some others that it
is the fundamental entities of the philosophers, abstract notions
expressed with the term ??ta.15) In the second part supporting the first major thesis (namely, 'noth-
ing is'), Gorgias, as the Anonymus author of De Melisso Xenophane
Gorgia (MXG) informs us, collected contradictory properties (t??a?t?a 979a 15) attributed by philosophers to ??ta; these contradictory prop-
erties, according to Anonymus, were discussed by Gorgias after the
'original proof (?et? t?? ?d??? a?t?? ?p?de???? 979a24), in which he
sought to show that 'it is not either for being or for not being'. In
the version given by the author of MXG these properties are dis-
cussed in 979b20-980a9, where we learn that 'if anything is' (e? d?
est??), it must be either generated or ungenerated, one or many, in motion or at rest.16) Each member of these pairs is proved impos- sible (with syllogisms which partly derive from axioms of the philoso-
phers themselves), and from that it is inferred that 'being is not'.
F. Romano (eds.), Gorgia e la Sofistica, Siculorum Gymnasium 38 (Catania 1985), 245-249; reprinted in: J. Mansfeld, Studies in the Histonography of Greek Philosophy (Assen 1990), 99-102.
15) A full discussion of this point is far beyond the scope of this paper and it does not affect my arguments here; however, I am inclined to follow Mansfeld (n. 14), 248-249, who has shown that ta d?ta in QNB are "the speculative theoretical constructs of the Presocratic philosophers, or the essential attributes of the things that are" (see also J. A. Palmer, Plato's Perception of Parmenides (Oxford 1999), 66 ff. and esp. 67 n. 24. G. B. Kerferd, Gorgias on nature or that which is not, Phronesis 1 (1955), 3-25, basically following G. Calogero, Studi sull' Eleatismo (Roma 1932), held that what is at stake in ONB is the "phenomenal world" (so H. J. Newiger, Unter- suchungen zu Gorgias' Schrifl ?ber das Nichtseiende (Berlin, New York 1973), 21-22).
16) The list of properties attributed to ??t? by philosophers is fuller in MXG 979b20- 980a9, for Sextus does not include the pair in motion/at rest. This pair is absent in the 'doxographical* summary in MXG 979al4-18, but it is traced in Xen. Mem. 1.1.14. (Mansfeld (n. 14), 246-247 investigates the historiography of philoso- phy as a Sophistic activity and he shows how later doxographical accounts depend on it.) Gorgias' ONB does not merely intend to record earlier ideas out of histor- ical interest, but it clearly seeks to refute them on the basis of 'logical* antinomies.
400 D. G. SPATHARAS
In Palamedes 25-26, the hero addresses the litigant; it is interest-
ing that he does not attack Odysseus on the grounds of personal characterization (27, although numerous adjectives may have been
used against Odyseeus).17) The characterization of the opponent is
built on two separate arguments: in the first one Palamedes explains that Odysseus relies his accusation upon belief (d??a), which is
defined as an ?p?st?tat?? p????a; serious accusations, we are told, should be based on firm knowledge.18) The second part of the apos-
17) T. Cole (n.8), 73 classifies the lack of ethos as one among other character- istics of late Fifth century rhetoric: "the absence of any attempt to give ethos to what is said by making it suggest the character of the person or class of person who is
saying it... points to the demands of the practice and demonstration text" (79). There is no doubt, I think, that Gorgias' Helen and Palamedes are intended for practical didactic purposes, and a need for general applicability is also discernible; but this is rather different from saying that ethos is totally absent. In his self-characterization, Palamedes lists his inventions, which are presented as a great benefaction to the Greeks and humanity in general (30). These are very unique virtues peculiar to this specific hero, and it is impossible to think of any of Gorgias' students who
might have been in a position to claim that he was the inventor of letters. It is now
true, that much of the self-characterization of Palamedes makes use of standard moral values (29-32): he has never been accused of anything before (29; notice that
???d???a, an inaccurate accusation, ??? e???sa? e?e????, is imputed to the oppo- nent); he does not cause pain to the elderly, he helps the young, he does not envy prosperous people, he sympathizes with those who suffer. .. (32): in short, Palamedes is in absolute conformity with moral standards and Gorgias is thus teaching prospec- tive rhetoricians how to use the stock of conventional morality. This does not, I am inclined to believe, imply the absence of ethos; it implies a twofold function of self-characterization: a) to present an impeccable Palamedes for the purposes of The Defence of Palamedes, that is a Palamedes whose profile does not deviate from the mythical account, and b) to offer a paradigm of general applicability. If my reading is correct, then Cole's suggestion seems to be an unjustified generalization (I acknowledge, of course, that Cole examines a greater range of texts; my point is simply that Gorgias is not among those who neglect ethos).
18) Gorgias' texts have regularly been interpreted under the light of a distinc- tion between 'knowledge' and 'belief; the most eloquent representative of this line of inquiry has been Kerferd (n. 1), 81-82, who claims that "it is possible to discern a common conceptual model... on the one hand is the real world, labelled truth or that which is true. The cognition of this real world is knowledge. But the com- monest cognitive state is opinion, not knowledge, and logos . . . operates upon opin- ion". A. Long, Methods of Argument in Gorgias' Palamedes (in: Voudouris, n. 5), 240, has rightly explained that "this is an unjustified systematization of Gorgias' prin- cipal preserved writings" (see also E. Schiappa, The Beginnings of Rhetorical Theory in Classical Greece (Yale 1999), 125-126. The artificiality of this distinction cannot be shown here; it will suffice to say that it appears only in Palamedes 3 and 24, in a context which does not allow for generalizations. Helen 11 has also been taken to
depict the prevalence of knowledge, but knowledge is not mentioned there at all. It is simply said that 'belief (d??a) is slippery (see D. M. MacDowell, Gorgias, Encomium of Helen (Bristol 1982) note ad loc). I consider that this systematization is
PATTERNS OF ARGUMENTATION IN GORGIAS 401
trophe to the opponent makes use of an argument from antinomy; what is at stake here is the unreliability of a litigant, who referring to the same person, attributes to it contradictory properties.
In this context, Palamedes refers to the speech of accusation in
which Odysseus had allegedly claimed that the defendant is both
wise and mad: wise in respect to his resourcefulness, mad in respect to the fact that he betrayed the Greeks (25):
?at?????sa? d? ??? d?a t?? e???????? ????? d?? ta ??a?t??tata, s?f?a? ?a? ?a??a?, ?spe? ??? ???? te t?? a?t?? ?????p?? e?e??. ?p?? ?e? ??? ?e
f?? e??a? te???e?t? te ?a? de???? ?a? p??????, s?f?a? ??? ?at????e??, ?p?? de ???e?? ?? p???d?d??? t?? ????da, ?a??a? . . .
What we have here is a game of chess played by the same player:
Gorgias, in defending Palamedes, has the privilege of answering accusations made by himself, while it should be noticed that these
accusations could reasonably have been put forward by one who
might have wished to capitalize on the overwhelming potential of
this hero; if Palamedes is so resourceful, then he has probably used
his resourcefulness for malicious purposes. The argument from antin-
omy, though more simple here than in OMB, is obvious; the 'two
totally contradictory properties' (d?? ta ??a?t??tata) ostensibly ascribed
to Palamedes by Odysseus make the latter's accusation contradic-
tory itself, from which it is logically inferred that his accusation is
unreliable (p?? ??? ??d?? t????tf p?ste?e??, ?st?? t?? a?t?? ????? ?????
p??? t??? a?t??? ??d?a? pe?? t?? a?t?? t? ??a?t??tata ???e?; Notice
the a?t??/??a?t??tata antithesis). The defendant concludes that
Odysseus is a liar (d?' a?f?te?a a? e??? ?e?d?? 26), by using the
same argument. He asks him if he deems wise men as f??????? or
a???t??; if wise men are a???t?? Odysseus' claim is a flagrandy untrue novelty. If they are f??????? they do not prefer sufferings
partly the concomitant of considering ONB as a treatise in which Gorgias' own theoretical credo is embedded and explicitly put forward. However, this text is (and probably was intended to be) open to different readings; far from putting for- ward new theories, it questions the validity of established ones. If the message of ONB is that philosophical systems claiming absolute approaches to truth are refutable, as I think it is, then it may turn out to be the worst source of information for Gorgias' own premises, if they existed at all. In my view, ONB should be read as a criticism on the process of philosophical reasoning, as a scrutiny of philosophical discourse. The fragility of philosophical reasoning reappears in Helen 13 as well.
402 D. G. SPATHARAS
to goods they already possess. The conclusion is: e? ??? ??? e???
s?f??, ??? ??a?t??? e? d' ??a?t??, ?? s?f?? e???; in either case Odysseus' accusations are proved to be false, the opponent is a liar.
Logical argumentation is thus in Gorgias' hands a means of bring-
ing out the ethos of the opponent; instead of a personal attack, we
are provided with an analysis of the logical contradictions resulting from the opponent's charges. Two points should be made: a) the
clarity with which this pattern of argumentation is presented serves
as an example of Gorgias' teaching practices and the need for
general applicability; having this example in mind, students of rhetoric
can easily argue from antinomy; b) the ethos of the opponent (not that of the defendant) is not presented on the basis of personal
attack; if one is reluctandy tempted to indulge in the discussion of
the morality of Gorgian rhetoric by comparing the tactic of attack
employed by Gorgias with that used by Fourth century orators (say
by Aeschines in Against Timarchos), where even false evidence is used
against the personalities of the opponents,19) then we may conclude
that Gorgias was almost naive.
3. Theonzation and examples
This pattern is related to reasoning involving speculation which is
not direcdy relevant to the theses defended. Criticism has focused
mainly on the discussion of logos by Gorgias in Helen 8-14; this part of the speech includes one of the earlier approaches to the func-
tion of speech and persuasion (pe???), especially in connection to
its impact on the human soul: poetry, incantations, the perception of speech by audiences, persuasive speech are all employed to exem-
plify the omnipotence of logos. However, it is critical to bear in mind
that Gorgias does not support his argumentation by the means of
theoretical discourse solely in this part of Helen. The final reason,
namely love, is from the very beginning of its analysis linked to
d??? ('vision'), and what follows is a theoretical evaluation of the
function of this sense in relation to the emotional world.
The relation of logos with psychological parameters has been
19) On inaccurate personal attack, see Halliwell, The Uses of Laughter in Greek
Culture, CQ,41 (1991), 292-294, P. Harding, Comedy and Rhetoric, in: I. Worthington (n. 1), 196-221.
PATTERNS OF ARGUMENTATION IN GORGIAS 403
admirably examined by Segal;20) what I intend to do is to show a) the ways in which the combination of theoretical generalization with
the use of examples contributes to the argumentation in Helen21) and b) that the separate arguments included in the discussion of
logos and love are underlined by a common pattern of analysis. In 8, logos is defined as a great ruler with extreme powers (?????
d???st?? ???a? est??); the realm of His activity is chiefly the emo-
tional world of men. In order to demonstrate the impact of logos
upon our emotions Gorgias brings in two examples corresponding to two different kinds of logoi: poetry, defined as ????? ????ta ??t???
(9), and incantations, e??e?? d?a ????? ?pfda? (10). Poetry awakens
within the souls of audiences emotions for the sufferings of others
(that is, with the 'suspension of disbelief audiences partake in the
reality of the literary event), and incantations?by means of magi- cal charming (???te?a? ?a? ?a?e?a?)?make the soul act independ-
endy of its own will. In both examples, speech enters the soul
physically, and the schema applied is common in both cases:
poetry ?? soul ?> emotions
incantations ?? soul ?? charming
Both poetry and verbal incantations (d?a ?????) enter the soul
(e?s???e, s??????????) and they affect it.22) The same pattern recurs in the analysis of love (15-19); the log-
ical discussion of a notoriously irrational emotion23) is foreshadowed
20) C. Segal, Gorgias and the Psychology of the Ugos, HSCPh 66 (1962), 99-155; according to this scholar Helen appeals more to the emotional aspect of persua- sion, whereas Palamedes makes use of logical reasoning. This distinction is somewhat elusive, because the rationalistic approach to logos is one thing, and the impact of logos upon the emotional world itself quite another (see I. Anastassiou (n. 5), 246-247).
21) Helen 8-14 can undoubtedly be regarded as an early piece of literary the- ory, and it is thus included in G. Lanata, Poetica Pre-Platonica, Testimonianze e Frammenti (Firenze 1962), and D. A. R?ssel - M. Winterbottom (eds.), Ancient Literary Cnticism (Oxford 1972); this aspect of Helen is beyond the scope of my paper.
22) A.-P. D. Mourelatos, ? G????a? ??a t? ?e?t?????a t?? ???ss?? (in: Voudouris ?. 5), 229-230 rightly maintains that the discussion of logos in Helen is basically behavioural, but he fails to observe that the same holds for the discussion of vision as well.
23) This is explicitly acknowledged by Gorgias (e? d' est?? ?????p???? ??s??a ?a? ????? ??????a, ??? ?? a???t??a ?e?pt??? ???* ?? at????a ????st??? 19); the superiority of love because of its divine nature (?? e? ??? ?e?? <?? e?e?> ?e?? ?e?a? d??a??? 19) is not the cornerstone of Gorgias' reasoning (it is mentioned in pass- ing), and it is worth noticing that love is rationalized in the way that divine pow- ers are rationalized in 6.
404 D. G. SPATHARAS
from the very beginning: Gorgias links ?'??? to d??? and he remarks
that 'the objects of our sight do not have the nature that we want
them to have, but the one they happen to have' (15). The prob-
lem, now, seems to be that 'soul is moulded by vision' (d?a d? t??
d?e?? ? ???? ?a? t??? t??p??? t?p??ta?). An example which supports this thesis is brought in:24) when soldiers face the weapon of the
enemies, their soul is in panic, so that they fly without considering the bad consequences of their action (16). The function of ???? is
based on the same pattern of analysis already employed in the con-
text of /ogoj-arguments:
vision ?? soul ?> flight (p?????a s??ata ...?? ?t??a?e t?? ????? ?> fe????s?? ??p?a???te?).
Vision is again coming physically (?????sa 16) into the soul,
which is 'moulded' (t?p??ta? 15; notice also that the images of
vision seen in the past are 'engraved', '?????a?e?' 17). The generalization about the function of vision is completed with
a reference to the emotions aroused by fine arts (19). It is assumed that
the function of painting and sculpture is to provide vision with pleas- ant images, and, what is more, from the products of art p???? and
???? can be generated. If one can possibly fall in love with a statue
(the example of Pygmalion is telling), then Helen's falling in love
with the statuesque body of Paris is perfecdy comprehensible. It has been made clear, I hope, that the reasoning in the dis-
cussion of both logos and love develops with a good deal of theo-
rization, which in some respects follows a common pattern of analysis
concerning the relation of the stimuli to the emotions that they invoke in the soul. But what is the value of this pattern in associ-
ation with the development of the reasoning for the case of Helen?
The answer is partly given by the text itself: at 12, Helen25) is
called a victim of persuasion, and at 19 we are told that we should
24) The examples adduced in 15-19 correspond to a distinction between nega- tive (16, 17, mainly fear) and positive (18, mainly pleasure invoked by painting and sculpture) emotions.
25) In spite of the textual problems the meaning is clear; my reading runs as follows: t?? ??? a?t?a ????e? ????sa? ?a? t?? ?????? ?p? ?????? ???e?? ?????? ??? ????sa? ?spe? e? ??a ??p?s??; ?p? ?a? t?? pe????? ??????? t? ????a* ?a?t?? pe???
a?????? e?d?? ??e? ?e? ??, t?? d? d??a??? t?? a?t?? ??e?. Gorgias is clearly trying to present Helen as a victim of persuasion. That he intends us to construe Helen as another example confirming the view that persuasion is as effective as violence and necessity is brought out by the wording itself (?a? t?? ??????).
PATTERNS OF ARGUMENTATION IN GORGIAS 405
not consider Helen's falling in love with Paris as a strange thing,
simply because her eye (sic) happened to see his body. Gorgias then
uses theorization because he relies on analogies: if logos is omnipo-
tent, as it is shown that it is, in what manner could Helen escape his power? If objects of vision contaminate our souls, as it is shown
that they do, then how could Helen's soul avoid contamination by the statuesque body of Paris? In the theoretical pattern of reason-
ing, the person defended is just another example that confirms the
theory. In addition, theorization has the virtue of explaining, giv-
ing logical meaning to things otherwise self-evident. Everyone has
perhaps fallen in love; but Gorgias is there to show (in his own way)
why and how this occurs. From an excuse love is elevated to a very
important reason.
In conclusion, the discussion of logos in Helen 8-14, however inter-
esting implications for the history of criticism in antiquity it may
have, is intended as a separate argument of equal significance; the-
orization is used both in the examination of logos and in that of love-
vision, where examples play an important role. The value of this
pattern lies in that, by generalizing, it is analogously applicable to
individual cases (in this case, Helen). We also have to assume, that
much of the persuasiveness of theorization through examples may have been the product of e?p????? experienced by audiences (or
readers) resulting from the impressive ability of the rhetor to apply elaborate 'knowledge' in demanding intellectual issues. There is one
more question: if logos is able to deceive, why should we become
the victims of the person who has just shown that logos deceives?
This is a matter of a second-order reading of Helen, and it should
be left open.26) Personal answers are, of course, always available.
4. Apagogie and the 'Russian doW argumentation
I include these two types in my classification reluctandy, for they concern formal schemata of reasoning. However, I hope that the
investigation of their role in Gorgias' reasoning will be compensating.
26) For the role of ap?t? in Gorgias, see Verdenius, Gorgias' Doctrine of Deception, in: G. B. Kerferd (ed.), The Sophists and their L?gacy (Hermes Einzelschriften 44; Wiesbaden 1981), 116-128; although Verdenius* study is learned and still up-to- date, in my view the phrase 'doctrine of deception' clearly overstates our evidence (see also n. 18).
406 D. G. SPATHARAS
Apagogie reasoning is employed both in Helen and in the dis-
cussion of motives in Palamedes', in the former, each reason is dealt
with separately, and none of them results from or presupposes the
preceding one.27) This is brought out from the text itself, because
the transition from each reason to the following one is clearly marked
(? ?a? ... 6 e? de ??a... 7, e? d? ????? ? pe?sa?... 8 ?a? dt? ?e?, e?
????... 15, t?? d? tet??t?? a?t?a? tf tet??tf ???f d???e??? 15). Gorgias invents four reasons, each one of which is intended to show the
same thing: Helen is not responsible. The same process is traced
in Palamedes 13-21. Gorgias distinguishes between two types of motives
(19): people commit crimes either in pursuit of a gain or in avoid-
ance of a loss (? ???d?? t? ?et???te? ? ????a? fe????te?) et tertium non
datur. All the motives presented fall within those two categories; the
arguments put forward show that if the defendant had committed
the crime of treason, he would have had the opposite results. The
apagogie reduction both in Helen and in Palamedes is used because
each independent argument is meant to be perceived as equally
strong as the rest of them. In other words, Gorgias is not com-
pelled to present each step in any particular order, because in these
cases his material does not impose upon him such a process. On the contrary, the first major division of Palamedes (6-12) pro-
ceeds with the 'Russian doll' schema, which I take it to be an indi-
cation of Gorgias' awareness that motives and actions corroborate
the argumentation in a different manner. As each new smaller doll
is brought out from a 'Russian doll', in the same manner each stage in the discussion of the actions preparing an alleged betrayal is pre- sented as logically following the preceding one.28) The more the
arguments represent a logical string of acts, the more the argu- mentation is benefited; this is why the first argument deals with
what should have normally been the starting point of a betrayal (6):
27) However, J. Porter, The Seductions of Gorgias, CA 12 (1993), 275, is certainly right in holding the view that "if Gorgias is trying to keep his aitiai apart, he is
trying no less hard to make that task next to impossible". 28) A. Long (n. 5), 235 remarks that "in a casual way examples of the technique
[sc. the 'Russian doll' or the 'Chinese box' as Long labels it] can be found in the Attic orators, but I know of nothing comparable to its use in the Palamedes"', accord-
ing to this scholar these examples are: Antiphon 3.4.6-7. 5.25, Lysias 9.8-10, Isocrates 21.7-8.
PATTERNS OF ARGUMENTATION IN GORGIAS 407
?de? ??? t??a p??t?? ????? ?e??s?a? t?? p??d?s?a?, ? de ???? ????? a?
e?? * p?? ?a? t?? ?e????t?? ????? a????? ?????? ????es?a? p??te???...
That each argument concedes the preceding one is indicated by the
introductory phrase of each new argument: ???a d? t??t? tf ???f d??at?? ?e??s?a? ... 7, ???a d? ?a? t??t? ?e??s??, ?a?pe? ??
?e???e??? ... 8, f?se? t?? . . . 9, ?a? d? t????? ?e??s?? ?a? ta ??
?e???e?a ... 11 etc. It is also worth our attention that the second
major division is actually introduced by conceding the first one: 'if
it were by all means possible (e? ????sta p??t?? ?d??????), for what
reason would I have wished to do these things?' (6). The value of
this type of argumentation lies in the fact that it presents a cohe-
sive string of arguments, based on logical assumptions. It can be
conceived as a representation of the crime, in which the defendant
is able to show that, what is presented by the accuser as a fact, is
nothing but assumption. The same pattern is followed in the philosophically oriented On
not Bang; each one of the three major theses concedes the previous one: nothing is (A), if it is, it is not possible to have knowledge of
it (B), if it L? and it is possible to have knowledge of it, it is impossible to
communicate it to others (C). Although the schema is the same, it
is not used to the same extent in the individual arguments of this
text, as it happens in Palamedes 6-12.
In short, Gorgias seems to arrange his arguments in accordance
to the nature of the case he defends; actions, normally developing in a linear, consecutive order suggest a similarly linear and exhaus-
tive representation,29) which demands some logical participation of
the audience. In the case of self-content arguments on the other
hand the argumentation is apagogL?; different theses are supported
by independent arguments: in Helen each reason is argued sepa-
rately, so that the refutation of her infamy is based on four equally
strong reasons; analogously, in Palamedes 13-22 each motive is dealt
with separately as well, and it is shown that none of them could
have led the hero to perform the act that he is accused of.
29) In Long's words (n. 18), 235 "the sequence of claims is assumed to be exhaustive, leaving the opponent no perch for any reply", and farther (237) he points out that the elimination of the defendant's opportunities "amount to ... a reconstruction of the alleged treachery from its beginning to its end".
408 D. G. SPATHARAS
Gorgias' argumentative process does not merely consist in prob-
abilities; Plato in his Phaedros (267a) simply singles out the type of
argument which makes Gorgias susceptible to criticism, in view of
the fact that probabilities do not reproduce factual reality. A sober
assessment of the argumentation used by this Sophist, which is based
on a close reading of his own preserved texts, shows that various
argumentative patterns are employed by Gorgias and that he is wise
enough not to ignore factual reality. Gorgias' reasoning is not as
simple as it is usually taken to be: I hope that I have removed from
the Sophist the infamy of probabilities, and that I have not engaged in the pa?????? of composing a G?????? encomium.
University of Glasgow, Dept. of Classics, G12 8QQ,